Back to Portfolio
COHR

Coherent Corp:

$243.29 USD 45.6B market cap 2026-03-27
Coherent Corp COHR BUFFETT / MUNGER / KLARMAN SUMMARY
1 SNAPSHOT
Price$243.29
Market CapUSD 45.6B
EVUSD 48.6B
Net DebtUSD 3.0B
Shares187.5M
2 BUSINESS

Coherent Corp is the world's broadest vertically integrated photonics company, developing optical transceivers, lasers, engineered materials, and semiconductor components for AI datacenters, industrial, and electronics markets. Formed through II-VI's $7B acquisition of Coherent Inc. in 2022, the company controls critical compound semiconductor platforms (InP, GaAs) that enable the optical interconnects inside every AI GPU cluster. Revenue is driven by hyperscaler demand for 800G/1.6T transceivers, with growing contributions from industrial lasers and Apple VCSELs.

Revenue: USD 5.81B Organic Growth: 23.4%
3 MOAT NARROW

Vertical integration from compound semiconductor epitaxy (InP, GaAs) through transceiver assembly. World's first 6-inch InP production platform provides cost and capacity advantages. 12-18 month customer qualification cycles create switching costs. Supply chain resilience with 20+ US manufacturing sites valued by hyperscalers. However, Chinese competitors are gaining share in commodity transceivers, and technology transitions (CPO, silicon photonics) could reset competitive positions.

4 MANAGEMENT
CEO: Jim Anderson (since Apr 2024)

Prioritizing debt paydown ($437M in FY2025), divesting non-core assets (A&D for $400M), heavy CapEx in AI-related capacity (InP 6-inch fabs, OCS production). No dividend. SBC ~$200M/year (~3.5% of revenue). Insider ownership only 0.5%.

5 ECONOMICS
17.8% (non-GAAP); 5.0% (GAAP) Op Margin
~4% (GAAP) ROIC
USD 0.19B FCF
2.5x Debt/EBITDA
6 VALUATION
FCF/ShareUSD 1.03
FCF Yield0.4%
DCF RangeUSD 170 - 220

Base case: 12% revenue CAGR FY2026-2031, operating margin expansion to 24%, 11% WACC, 3% terminal growth. Bull case ($280-340) assumes 15% CAGR + 27% margins. Bear case ($90-130) assumes flat revenue + margin stagnation.

7 MUNGER INVERSION -32.6%
Kill Event Severity P() E[Loss]
AI datacenter spending slowdown/pause -40% 25% -10.0%
Transceiver pricing pressure / commoditization from Chinese competitors -25% 30% -7.5%
Customer concentration loss (2 customers >10% of revenue) -30% 20% -6.0%
Technology disruption (silicon photonics, LPO replaces EML) -35% 15% -5.3%
Competition from Broadcom, Lumentum, InnoLight intensifies -15% 25% -3.8%

Tail Risk: In a severe AI winter scenario combined with Chinese competition capturing share and technology disruption, COHR could revert to its FY2023 trough valuation ($35-45 range). Net debt of $3.0B would become burdensome if EBITDA contracts. The 2022 acquisition created $6.8B in goodwill that could require impairment.

8 KLARMAN LENS
Downside Case

AI datacenter CapEx pauses as hyperscalers reassess ROI on AI investments. Transceiver ASPs decline 20-30% as Chinese manufacturers gain 800G share. Revenue stalls at $5.5-6.0B, margins compress. Stock reverts to $90-130.

Why Market Wrong

The market may be underestimating (1) the duration and magnitude of the AI datacenter build cycle, (2) the difficulty of replicating Coherent's vertical integration in compound semiconductors, and (3) the margin expansion potential as the portfolio optimization removes low-margin businesses.

Why Market Right

The market may be correctly pricing in multi-year AI growth but ignoring (1) cyclicality risk inherent in semiconductor/optical equipment, (2) the gap between GAAP and non-GAAP results ($200M+ SBC is real dilution), (3) execution risk on margin targets with only 1 year of track record under new CEO.

Catalysts

Positive: 1.6T transceiver volume ramp (H2 CY2026), OCS production scale-up, Apple VCSEL revenue restart (H2 CY2026), continued debt paydown. Negative: AI CapEx cuts by any major hyperscaler, tariff escalation, InnoLight gaining >30% share in 800G.

9 VERDICT WAIT
B T3 Adaptable
Strong Buy$130
Buy$170
Sell$320

Coherent Corp is a genuinely high-quality photonics business riding the structural AI datacenter buildout. Vertical integration in InP/GaAs compounds is a real competitive advantage. However, at $243 (38x forward non-GAAP P/E, 0.4% FCF yield), the stock prices in near-flawless execution on margin expansion and continued AI demand acceleration. GAAP profitability is poor (ROE 0.6%, 3.3% FCF margin). Wait for a pullback to $170 range where the growth story is priced at a reasonable premium with margin of safety.

🧠 ULTRATHINK Deep Philosophical Analysis

COHR - Ultrathink Analysis

The Real Question

The question is not "will AI datacenters keep getting built?" -- they will. The real question is: can Coherent Corp convert its current position as the broadest photonics platform into a durable, high-return business, or is it destined to remain a capital-intensive middleman in a supply chain where the real value accretes to the chip designers (NVIDIA) and the hyperscaler end customers?

This is the central tension. Coherent makes the plumbing that connects AI chips to each other. Without optical transceivers, GPU clusters are deaf, mute islands of silicon. The transceivers are essential. But "essential" and "valuable" are not synonyms. Water is essential; water utilities earn mediocre returns. The question is whether Coherent's vertical integration in compound semiconductors creates enough differentiation to earn returns meaningfully above its cost of capital over a full cycle, or whether the optical transceiver market will follow the grim trajectory of most hardware commodities.

Hidden Assumptions

The market is making three assumptions that deserve scrutiny:

Assumption 1: AI CapEx is a permanent step-change, not a cycle. The bull narrative treats hyperscaler capital spending as a one-way escalator. History suggests otherwise. Cloud computing spending surged in 2020-2021, then contracted sharply in 2022-2023, causing inventory corrections across the optical supply chain. Coherent's own FY2024 results (revenue down 9%, Networking profit down 24%) demonstrate this cyclicality. The current belief that "this time it's different because AI" may prove correct, but the burden of proof should be on the bulls. Every infrastructure build cycle in technology history -- fiber optic, 3G/4G/5G, cloud -- has experienced painful mid-cycle corrections.

Assumption 2: Vertical integration provides durable pricing power. Coherent's InP and GaAs fabs are genuinely difficult to replicate. But difficulty of replication and pricing power are not the same thing. Intel had the world's most advanced semiconductor fabs for decades and still saw its pricing power erode as competition caught up and architecture shifts (ARM, chiplets) changed the game. InnoLight is already a $2B+ transceiver company growing faster than Coherent in certain segments, with lower cost structure. The 6-inch InP platform is a real advantage, but it's a process advantage that can be copied given enough capital and time.

Assumption 3: The GAAP/non-GAAP gap is temporary. Management presents $3.53 non-GAAP EPS as the "true" earnings while reporting -$0.52 GAAP EPS. The $200M+ annual stock-based compensation is a real cost to shareholders -- it dilutes ownership every single year. The $350-400M in amortization of acquired intangibles reflects a real cost: the $7B paid for the Coherent acquisition. Adjusting these away is convenient for the earnings story but misleading about actual returns on invested capital.

The Contrarian View

For the bears to be right, the following would need to be true:

  1. AI datacenter spending growth decelerates from 40%+ to single digits within 2-3 years, as hyperscalers realize that inference workloads require less incremental optical bandwidth than training clusters.

  2. Chinese transceiver manufacturers achieve 800G and 1.6T parity by 2027-2028, compressing transceiver ASPs by 30-40% and making vertical integration a cost disadvantage rather than advantage (because vertically integrated fabs have higher fixed costs).

  3. Silicon photonics (led by Broadcom) emerges as the dominant platform for next-generation interconnects, rendering Coherent's compound semiconductor investments partially stranded.

  4. The company never achieves its long-term margin targets (42% gross, 24% operating) because each generation of transceivers requires expensive new capacity that resets the margin improvement cycle.

None of these are far-fetched. The bear case is not some apocalyptic scenario -- it is the base case for most hardware suppliers throughout technology history.

Simplest Thesis

Coherent is a bet that photons will be as important as electrons in the AI era, and that the company that controls the most critical materials and manufacturing processes for manipulating photons will earn premium returns -- but only if bought at a price that accounts for the cyclical and competitive risks inherent in hardware manufacturing.

Why This Opportunity Exists

The opportunity -- such as it is -- exists because of a fundamental tension between the stock's narrative and its economics:

The narrative is irresistible: AI is the most transformative technology since the internet, datacenters are being built at unprecedented scale, every GPU needs optical transceivers, and Coherent is the most vertically integrated supplier. This narrative has driven the stock from $45 to $300 in 12 months.

The economics are sobering: 0.6% ROE, 3.3% FCF margin, $3B net debt, and a business model that requires perpetual heavy investment just to maintain competitive position. The company spent $441M in CapEx in FY2025 to generate $193M in free cash flow.

The opportunity for a patient investor exists precisely because the narrative will eventually collide with the economics. When AI CapEx growth inevitably decelerates (from 50%+ to perhaps 15-20%), the stock will likely correct significantly from its current premium valuation. At that point -- perhaps $130-170 -- an investor would be buying a legitimate technology franchise at a price that reflects its actual cash generation capability rather than its aspirational narrative.

The mispricing, in other words, is not that the market is wrong about Coherent's competitive position. The market is correct that this is a well-positioned company. The mispricing is that the market is paying today's price for tomorrow's earnings while ignoring the cyclicality that has characterized this business in every prior period.

What Would Change My Mind

Three specific observations would convert this from WAIT to BUY even at current prices:

  1. GAAP operating margin exceeding 15% for two consecutive quarters, proving that the non-GAAP adjustments are genuinely temporary rather than structural.

  2. FCF exceeding $600M on a trailing twelve-month basis, demonstrating that the CapEx investment cycle is yielding genuine free cash flow rather than perpetually absorbing it.

  3. Achieving 42% gross margin on a non-GAAP basis while growing revenue, proving that the margin expansion is driven by pricing power and mix rather than just cost cutting.

Conversely, one observation would make the bear case definitive: if InnoLight or another Chinese competitor ships 1.6T transceivers at parity quality with Coherent at 20%+ lower prices, the vertical integration story collapses from competitive advantage to competitive albatross.

The Soul of This Business

At its core, Coherent is a materials science company. It grows crystals -- indium phosphide, gallium arsenide, silicon carbide -- and turns them into devices that manipulate light. This has been the soul of the business since II-VI was founded in 1971. The company's deepest competitive advantage is not its end products (transceivers, lasers, VCSELs) but its mastery of the materials beneath them.

This matters because materials expertise compounds slowly but durably. You cannot learn to grow InP crystals to the required quality by throwing money at the problem for two years. It takes decades of accumulated process knowledge, yield improvement, and materials characterization. This is why Coherent's position is more defensible than a surface-level analysis of the transceiver market would suggest.

But it also creates a fragility. Materials companies are inherently capital-intensive. They earn their returns through manufacturing excellence, not through network effects or switching costs. When the cycle turns, there is no flywheel to sustain growth -- just fixed costs that become a burden rather than a lever.

The soul of Coherent is the tension between the permanence of its materials knowledge and the cyclicality of the markets it serves. This is a company that will still exist and still matter in twenty years. The question is whether shareholders who buy at $243 will earn adequate returns over those twenty years, given the inevitable valleys between the peaks. History, and Graham's margin of safety principle, suggest patience is the wiser course.

Executive Summary

Three-Sentence Thesis

Coherent Corp is the world's broadest and deepest vertically integrated photonics company, riding the structural AI datacenter buildout wave through its optical transceivers, EML lasers, and indium phosphide semiconductor platforms. The company's transformation from a diversified industrial conglomerate (II-VI) into a focused photonics growth story is accelerating under CEO Jim Anderson's portfolio optimization strategy, with Networking segment revenues growing 49% YoY in FY2025 to $3.4B. However, at $243/share (forward P/E ~38x on non-GAAP, trailing P/E 269x on GAAP), the stock prices in substantial execution on long-term margin expansion targets (42% gross, 24% operating) and continued AI datacenter demand acceleration, offering limited margin of safety for a value investor.

Key Metrics Dashboard

Metric Value Assessment
Revenue (FY2025) $5.81B (+23% YoY) Strong growth
GAAP Net Income -$80.6M (loss) Impaired by restructuring
Non-GAAP EPS $3.53 (+191% YoY) Rapid improvement
GAAP EPS (TTM) $1.01 Barely profitable on GAAP
FCF (FY2025) $193M Modest, CapEx-heavy
FCF Yield 0.4% Very low
Net Debt ~$3.0B Elevated, deleveraging
ROE (GAAP) 0.6% Far below Buffett threshold
Gross Margin (non-GAAP FY2025) 37.9% Improving, target 42%
Operating Margin (non-GAAP FY2025) 17.8% Improving, target 24%
P/E (Forward, non-GAAP) ~38x Growth premium
EV/EBITDA 46.6x Expensive

Recommendation: WAIT - Monitor for pullback to $140-170 range


Phase 0: Business Understanding

What Does Coherent Do?

Coherent Corp is a vertically integrated photonics company headquartered in Saxonburg, Pennsylvania. It develops, manufactures, and markets lasers, optical transceivers, engineered materials, and optoelectronic components. The company was formed through II-VI Incorporated's $7B acquisition of the original Coherent Inc. in 2022, creating the world's most comprehensive photonics platform.

Three Reporting Segments (FY2025):

Segment Revenue Profit Margin Key Products
Networking $3,421M (59%) $644M 18.8% Optical transceivers, EMLs, CW lasers, CPO components, OCS
Lasers $1,435M (25%) $317M 22.1% Excimer lasers, industrial lasers, display/semi cap equipment
Materials $954M (16%) $355M 37.2% SiC substrates, VCSELs, thermoelectrics, engineered materials

Revenue by Market (FY2025):

  • Communications: ~$3.4B (59%) - Datacenter + Telecom
  • Industrial: ~$1.5B (26%) - Precision manufacturing, semi cap, display, aerospace
  • Electronics: ~$0.5B (9%) - Consumer electronics (VCSELs/Apple), automotive (SiC), wireless
  • Instrumentation: ~$0.4B (7%) - Life sciences, scientific research

Why It Matters for AI Infrastructure

Coherent is a critical supplier in the AI datacenter optical interconnect value chain:

  1. Optical Transceivers (800G, 1.6T, 3.2T roadmap): Convert electrical signals to optical signals for datacenter networking. Every AI GPU cluster needs thousands of transceivers.
  2. EML Lasers (Electro-absorption Modulated Lasers): The key semiconductor laser inside high-speed transceivers. Coherent makes these internally on indium phosphide.
  3. CW Lasers: Essential for co-packaged optics (CPO), the next-generation optical interconnect approach.
  4. Optical Circuit Switches (OCS): New product line enabling reconfigurable datacenter networks. $2B TAM expansion.
  5. VCSELs: Vertical-cavity surface-emitting lasers for 3D sensing (Apple partnership) and short-range datacenter links.

Vertical Integration Advantage

Coherent's deepest moat comes from vertical integration:

  • InP epitaxy (6-inch wafer platform, industry first) -> EML and CW laser chips
  • GaAs epitaxy (6-inch) -> VCSEL arrays
  • Compound semiconductor fabs in US and Europe
  • Transceiver assembly and test facilities
  • Optical components (lenses, filters, gratings)

This means Coherent controls the most critical components inside its transceivers, reducing cost, improving supply chain resilience, and creating barriers to entry.


Phase 1: Risk Analysis (Inversion - "How Could This Destroy Us?")

Risk Register

# Risk Event P(Event) Impact Expected Loss Timeframe
1 AI datacenter spending slowdown/pause 25% -40% -10.0% 12-24 months
2 Technology disruption (silicon photonics, LPO) 15% -35% -5.3% 2-5 years
3 Transceiver pricing pressure / commoditization 30% -25% -7.5% 12-36 months
4 Customer concentration risk (2 customers >10%) 20% -30% -6.0% Ongoing
5 Integration/restructuring execution failure 15% -20% -3.0% 12-24 months
6 Tariff/trade war escalation (China exposure) 20% -15% -3.0% 6-18 months
7 Debt burden constraining investment 10% -20% -2.0% 12-36 months
8 SiC/automotive demand secular decline 25% -10% -2.5% 12-36 months
9 Management execution risk (new CEO) 10% -15% -1.5% 12-24 months
10 Competition from Broadcom, Lumentum, InnoLight 25% -15% -3.8% Ongoing

Total Expected Downside: -44.6% (non-additive, correlated risks) Realistic Correlated Downside: -30% to -40% in bear scenario

Detailed Risk Analysis

1. AI Datacenter Spending Slowdown (Highest Impact) The entire bull case for COHR rests on continued massive AI infrastructure buildout. Networking revenue grew 49% YoY in FY2025, driven by hyperscaler demand. If AI CapEx spending pauses (as happened with cloud spending in 2022-23), COHR would face severe revenue headwinds. Management noted 2 customers each >10% of revenue. Hyperscaler CapEx commitments for 2026 remain elevated, but any macro recession or AI monetization skepticism could trigger cuts.

2. Technology Disruption Optical interconnect technology is evolving rapidly. Key risks:

  • Silicon photonics from Broadcom, Intel could commoditize transceivers
  • Linear-drive pluggable optics (LPO) could reduce EML laser requirements
  • Co-packaged optics (CPO) transition - COHR is positioning for this but timing uncertain
  • 3.2T standard still in development; COHR's 400G/lane differential EML is promising but unproven at scale

3. Pricing Pressure Chinese transceiver manufacturers (InnoLight, Eoptolink) are gaining share in 400G/800G. While COHR competes on vertical integration and supply chain resilience (especially US manufacturing), Chinese competitors offer 20-30% price advantages. The recent tariff environment may benefit COHR near-term but creates long-term uncertainty.

4. Customer Concentration Two customers exceeded 10% of FY2025 revenue (likely Meta and Microsoft/Google). Loss of any major hyperscaler account would be devastating. Apple renewed a multiyear VCSEL agreement, but the prior Apple design change caused a $270M revenue decline in FY2024.

5. Integration Complexity The II-VI/Coherent merger created a sprawling enterprise (30,000 employees, 20+ US manufacturing sites). Multiple restructuring plans are ongoing:

  • 2023 Plan: $53M charges in FY2025
  • 2025 Plan: $107M charges in FY2025
  • Synergy Plan: $17M charges in FY2025
  • Assets held-for-sale impairments: $85M
  • Total restructuring/impairment: $262M in FY2025 alone

Phase 2: Financial Analysis

Revenue Analysis (5-Year)

FY Revenue YoY Growth Communications Industrial Electronics Instrumentation
2021 $3.11B - - - - -
2022 $3.32B +6.8% - - - -
2023 $5.16B +55.4% $2.3B* $1.6B* $0.9B* $0.5B*
2024 $4.71B -8.7% $2.3B $1.5B $0.6B $0.4B
2025 $5.81B +23.4% $3.4B $1.5B $0.5B $0.4B

FY2023 includes full year of Coherent Inc. acquisition (closed Jul 2022). FY2021-2022 are legacy II-VI only.

5-Year Revenue CAGR: 11.9% (artificially inflated by acquisition) Organic growth FY2024-2025: 23.4% (strong, driven by AI)

Profitability Analysis

Metric FY2021 FY2022 FY2023 FY2024 FY2025
Gross Margin (GAAP) 37.9% 38.2% 31.4% 30.9% 35.4%
Gross Margin (non-GAAP) - - 34.2% 34.3% 37.9%
Operating Margin (GAAP) 12.9% 3.5% -0.7% 2.0% 5.0%
Operating Margin (non-GAAP) - - 13.1% 13.1% 17.8%
Net Margin (GAAP) 8.4% 7.1% -5.0% -3.3% 0.8%
ROE (GAAP) 5.7% 3.4% -5.6% -3.7% 0.6%

Key Observation: Massive gap between GAAP and non-GAAP results. Non-GAAP excludes:

  • Share-based compensation (~$200M/year)
  • Amortization of acquired intangibles (~$350-400M/year)
  • Restructuring charges ($160M in FY2025)
  • Asset impairments ($85M in FY2025)

This is a company where GAAP profitability is genuinely poor. The $6.8B goodwill from the II-VI/Coherent merger creates massive amortization drag. On a GAAP basis, ROE has never exceeded 6% post-merger and was negative for FY2023-2024.

Balance Sheet

Metric FY2022 FY2023 FY2024 FY2025
Total Assets $7.8B $13.7B $14.5B $14.9B
Goodwill + Intangibles ~$3.0B ~$8.0B ~$7.5B ~$7.0B
Total Debt $2.4B $4.5B $4.3B $3.9B
Cash $2.6B $0.8B $0.9B $0.9B
Net Debt $0.1B $3.7B $3.4B $3.0B
Total Equity $4.4B $7.2B $7.6B $8.1B
D/E Ratio 0.79 0.90 0.86 0.79
Net Debt/EBITDA 0.1x 4.1x 3.3x 2.5x

Deleveraging trajectory is positive. Net Debt/EBITDA declined from 4.1x to 2.5x. The A&D sale ($400M proceeds) will further reduce debt. However, $3.0B in net debt with $0.9B cash is still a significant burden, and the ~$200M annual interest expense is material.

Cash Flow Analysis

Metric FY2021 FY2022 FY2023 FY2024 FY2025
Operating CF $574M $413M $634M $546M $634M
CapEx -$146M -$314M -$436M -$347M -$441M
FCF $428M $99M $198M $199M $193M
FCF Margin 13.8% 3.0% 3.8% 4.2% 3.3%
Debt Paydown - - - $234M $437M

FCF is disappointing for a $45B market cap company. $193M FCF on $5.8B revenue = 3.3% FCF margin. The company is investing heavily in capacity expansion (InP 6-inch fabs, VCSEL capacity for Apple, OCS production). FCF yield at current price is only 0.4% ($193M / $45.6B market cap).

Owner Earnings Calculation (Buffett Method)

Net Income (GAAP):                    $30M
+ D&A (incl. intangible amort):      ~$800M
+ Restructuring (non-recurring):      $160M
+ Impairments:                        $85M
- Maintenance CapEx (estimated 60%):  -$265M
- Stock-Based Comp:                   -$200M
= Approximate Owner Earnings:         ~$610M

Owner earnings per share: ~$3.25 (closer to non-GAAP EPS of $3.53) Owner earnings yield: 1.3% at $243/share

Valuation

Forward P/E Analysis:

  • Non-GAAP FY2026E EPS: ~$5.50-6.50 (assuming continued revenue growth + margin expansion)
  • Forward P/E at $243: ~37-44x non-GAAP
  • GAAP P/E: 269x (trailing) - meaningless due to non-recurring items

DCF Model (Base Case):

Assumptions:
- Revenue FY2026: $6.5B (+12%)
- Revenue CAGR FY2026-2031: 12% (AI + industrial)
- Terminal growth: 3%
- Non-GAAP operating margin expansion: 17.8% -> 24% by FY2029
- Discount rate (WACC): 11%
- Share count: 190M (slight dilution from SBC)

DCF Fair Value: $170-220/share
Current Price: $243.29
Premium to Fair Value: +10-43%

DCF Model (Bull Case - Execution on All Targets):

Assumptions:
- Revenue FY2026: $7.0B (+20%)
- Revenue CAGR FY2026-2031: 15% (accelerating AI + CPO + OCS)
- Non-GAAP operating margin: 17.8% -> 27% by FY2030
- Discount rate: 10%

DCF Fair Value: $280-340/share

DCF Model (Bear Case - AI Slowdown):

Assumptions:
- Revenue FY2026: $5.8B (flat)
- Revenue CAGR FY2026-2031: 5%
- Non-GAAP operating margin stalls at 18%
- Discount rate: 12%

DCF Fair Value: $90-130/share

Relative Valuation:

Peer EV/EBITDA Forward P/E Revenue Growth
COHR 46.6x 38x +23%
Broadcom (AVGO) 30x 28x +44%
Lumentum (LITE) 35x 30x +15%
Fabrinet (FN) 25x 25x +15%
Ciena (CIEN) 20x 22x +10%

COHR trades at a significant premium to most optical networking peers, justified only by its higher growth rate and vertical integration story.


Phase 3: Moat Analysis

Moat Assessment: NARROW (Widening)

Moat Sources:

Source Strength Evidence Durability
Vertical Integration Strong InP/GaAs epitaxy, 6-inch wafer platform, internal EML production 10+ years
Switching Costs Moderate Qualification cycles 12-18 months, supply chain resilience premium 5-10 years
Technology Leadership Strong First 6-inch InP production, 400G/lane differential EML, OCS 3-5 years (must stay ahead)
Scale Moderate 30,000 employees, 20+ US sites, global footprint 10+ years
Customer Relationships Moderate Apple multiyear VCSEL deal, hyperscaler qualifications 3-5 years per contract

Moat Width: NARROW - While the vertical integration is genuinely difficult to replicate, the competitive landscape in optical transceivers is intense. Chinese manufacturers are catching up in 800G, and the transition to 1.6T/3.2T could reset competitive positions. Coherent's moat is widening due to InP capacity investments, but it's not yet a wide moat business.

Pricing Power Test: Mixed. Non-GAAP gross margins improved from 34.3% to 37.9% in FY2025, suggesting some pricing power. However, management explicitly cited "pricing optimization" alongside "cost reductions" - this is a company that needs to fight for margins, not one that can simply raise prices.

Key Moat Risks

  1. Chinese competition in commodity transceivers (InnoLight is growing rapidly)
  2. Broadcom's silicon photonics approach could commoditize discrete component-based transceivers
  3. Technology transitions (CPO, LPO) could strand existing capacity investments
  4. Apple dependency - the VCSEL relationship is important but Apple has shown willingness to change suppliers (FY2024 design change)

Phase 4: Decision Synthesis

Management Assessment

CEO: Jim Anderson (since April 2024, ~2 years)

  • Former President of Lattice Semiconductor
  • Driving portfolio optimization strategy (selling A&D business, restructuring)
  • Background in semiconductor industry leadership
  • Too early to fully assess track record at Coherent

CFO: Sherri Luther (since before merger)

  • Continuity through the II-VI/Coherent integration
  • Overseeing successful deleveraging ($437M debt paydown in FY2025)

Capital Allocation:

  • Prioritizing debt paydown (good - reduces risk)
  • Heavy CapEx investment in AI-related capacity (InP 6-inch, OCS)
  • A&D divestiture for $400M (strategic simplification)
  • No dividend (appropriate given debt levels)
  • SBC is substantial (~$200M/year, ~3.5% of revenue)
  • Insider ownership: 0.5% (low - typical for large-cap)

Buffett Quality Checklist

Criterion Pass/Fail Notes
Simple business? Partial Photonics is complex, but datacenter story is clear
Profitable 10+ years? FAIL Only 2 years of GAAP profitability post-merger
Consistent FCF? FAIL FCF margins 3-4%, CapEx intensive
ROE > 15%? FAIL GAAP ROE: 0.6% (FY2025)
D/E < 0.5? FAIL D/E: 0.79
Management skin in game? Weak 0.5% insider ownership
Identifiable moat? Partial Narrow moat, widening
Margin of safety at current price? FAIL Trading above DCF base case

Buffett Score: 1/8 - This is emphatically NOT a Buffett-style investment. It fails virtually every quality gate for traditional value investing.

Why This Is Interesting Despite Failing Value Tests

  1. Structural demand driver - AI datacenter buildout is real and multi-year
  2. Margin expansion trajectory is credible (37.9% -> 42% gross, 17.8% -> 24% operating)
  3. Vertical integration is a genuine competitive advantage being strengthened
  4. Portfolio optimization is removing drag from non-core businesses
  5. Deleveraging is progressing well ($437M debt reduction in FY2025)
  6. New product cycles (1.6T transceivers, OCS, CPO) provide multi-year growth vectors

Leopold Aschenbrenner / Situational Awareness Context

Aschenbrenner's Situational Awareness fund holds COHR at 2.1% weight ($89M) as part of an AGI infrastructure thesis. This aligns with the belief that AI compute will scale dramatically over the next 3-5 years, requiring massive optical interconnect build-out. Key reasoning:

  • Every AI GPU cluster needs thousands of optical transceivers
  • As model sizes grow, more inter-GPU communication = more optical bandwidth
  • COHR is the most vertically integrated provider in this supply chain
  • The transition from 800G -> 1.6T -> 3.2T provides a multi-year upgrade cycle

However, this is a momentum/growth thesis, not a value thesis. The position sizing (2.1%) reflects it as a supporting position, not a high-conviction anchor.

Entry Price Framework

Level Price Forward P/E Justification
Strong Buy $130 ~20x non-GAAP Near bear-case DCF; significant margin of safety
Accumulate $170 ~27x non-GAAP Base-case DCF midpoint; reasonable for growth
Current $243 ~38x non-GAAP Priced for bull-case execution
Sell $320 ~50x non-GAAP Excessive euphoria pricing

Monitoring Triggers

Trigger Action Current Level
Networking revenue growth < 20% YoY Reassess thesis 49% (healthy)
Non-GAAP gross margin < 35% Reassess thesis 37.9% (improving)
Net Debt/EBITDA > 3.0x Reduce conviction 2.5x (improving)
Loss of top-2 customer Exit None reported
1.6T transceiver ramp delay Monitor On track (initial revenue Q4 FY2025)
CPO transition disrupting pluggable Reassess thesis Not yet material

Verdict

WAIT - Strong Business, Excessive Valuation

Coherent Corp is a genuinely high-quality photonics business with real competitive advantages in a structurally growing market. The AI datacenter optical interconnect opportunity is real, the vertical integration strategy is sound, and management is executing well on portfolio optimization and margin expansion.

However, the stock is priced for perfection. At $243, investors are paying:

  • 269x trailing GAAP P/E
  • 38x forward non-GAAP P/E
  • 46.6x EV/EBITDA
  • 0.4% FCF yield

For a company that:

  • Has never posted GAAP ROE above 6% post-merger
  • Generates only 3.3% FCF margins
  • Still carries $3.0B in net debt
  • Has 0.5% insider ownership
  • Faces intense competition from Chinese transceiver manufacturers

The risk-reward is unfavorable at current prices. A 30% pullback to the $170 range would create a much more compelling entry point where the AI growth story is priced at a reasonable premium rather than requiring flawless execution.

Position Sizing (if entered at accumulate price): 2-3% portfolio weight. This is a high-beta (1.91), cyclical technology stock with significant execution risk. Not appropriate for concentrated positions.

Quality Grade: B - Good business with narrow-widening moat, but poor GAAP profitability, heavy debt, and acquisition integration challenges prevent higher rating.

Tier: T3 Adaptable - Growth company in technology transition, requires continuous adaptation and investment to maintain competitive position.